Multivariate Prediction of PM10 Concentration by LSTM Neural Networks

Air presence of particulate pollutants is an environmental problem with significant health issues. Monitoring their concentration is a key factor for the correct management of urban activities. In the smart cities scenario, the most fruitful tools for such application are sensor networks combined with machine learning techniques. In this work, neural networks are employed to forecast particulate concentration of air pollutants using a novel multivariate approach. We analyzed five years of data relating to PM10 concentration, studying the performance of different models based on the Long Short Term Memory paradigm, optimizing their hyperparameters accordingly. The tests show good results in terms of approximation and generalization capabilities, along with a sensible dependence on the weather conditions.

[1]  Boen Zhang,et al.  Influences of wind and precipitation on different-sized particulate matter concentrations (PM2.5, PM10, PM2.5–10) , 2018, Meteorology and Atmospheric Physics.

[2]  Antonello Rizzi,et al.  Refining accuracy of environmental data prediction by MoG neural networks , 2003, Neurocomputing.

[3]  Massimo Panella,et al.  Embedding of time series for the prediction in photovoltaic power plants , 2016, 2016 IEEE 16th International Conference on Environment and Electrical Engineering (EEEIC).

[4]  Frank de Leeuw,et al.  Air quality status and trends in Europe , 2014 .

[5]  ENVIRONMENTAL OUTLOOK TO 2050 : The consequences of Inaction Key Findings on Health and Environment , 2022 .

[6]  W. Ouyang,et al.  The washing effect of precipitation on particulate matter and the pollution dynamics of rainwater in downtown Beijing. , 2015, The Science of the total environment.

[7]  Massimo Panella,et al.  Selection of clinical features for pattern recognition applied to gait analysis , 2017, Medical & Biological Engineering & Computing.

[8]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[9]  B. Tomassetti,et al.  Mapping of Alternaria and Pleospora concentrations in Central Italy using meteorological forecast and neural network estimator , 2013, Aerobiologia.

[10]  Massimo Panella,et al.  A Neural Network Based Prediction System of Distributed Generation for the Management of Microgrids , 2019, IEEE Transactions on Industry Applications.

[11]  WHO Air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide , 2006 .

[12]  Marcella Busilacchio,et al.  Recursive neural network model for analysis and forecast of PM10 and PM2.5 , 2017 .

[13]  D. Poyner,et al.  Temperature and humidity effects on particulate matter concentrations in a sub-tropical climate during winter , 2017 .

[14]  Massimo Panella,et al.  2-D Convolutional Deep Neural Network for Multivariate Energy Time Series Prediction , 2019, 2019 IEEE International Conference on Environment and Electrical Engineering and 2019 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe).

[15]  H. Herrmann,et al.  Trans-boundary PM10: Quantifying impact and sources during winter 2016/17 in eastern Germany , 2019, Atmospheric Environment.

[16]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[17]  Massimo Panella,et al.  Shapes classification of dust deposition using fuzzy kernel-based approaches , 2016 .